cs.RO（2025-09-30）

📊 共 29 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱一：机器人控制 (Robot Control) (22 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (3) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱三：空间感知与语义 (Perception & Semantics) (2)

🔬 支柱一：机器人控制 (Robot Control) (22 篇)

#题目一句话要点标签🔗⭐
1 OmniRetarget: Interaction-Preserving Data Generation for Humanoid Whole-Body Loco-Manipulation and Scene Interaction OmniRetarget：交互保持的人形机器人全身运动操作与场景交互数据生成。 humanoid humanoid robot locomotion
2 A Systematic Study of Large Language Models for Task and Motion Planning With PDDLStream 系统性研究大语言模型在基于PDDLStream的任务与运动规划中的应用 motion planning large language model task and motion planning
3 MLA: A Multisensory Language-Action Model for Multimodal Understanding and Forecasting in Robotic Manipulation 提出MLA多感官语言-动作模型，增强机器人操作中的多模态理解与预测能力 manipulation world model vision-language-action
4 Reinforced Embodied Planning with Verifiable Reward for Real-World Robotic Manipulation REVER：基于可验证奖励的强化具身规划，用于真实世界机器人操作 manipulation physically plausible embodied AI
5 ISyHand: A Dexterous Multi-finger Robot Hand with an Articulated Palm ISyHand：一种具有铰接掌部的灵巧多指机器人手 humanoid humanoid robot manipulation
6 Learning Human Reaching Optimality Principles from Minimal Observation Inverse Reinforcement Learning 提出基于最小观测逆强化学习的人体手臂运动最优性建模方法 humanoid humanoid robot reinforcement learning
7 SAC Flow: Sample-Efficient Reinforcement Learning of Flow-Based Policies via Velocity-Reparameterized Sequential Modeling 提出SAC Flow算法，通过速度重参数化建模，提升Flow-based策略在强化学习中的样本效率。 manipulation reinforcement learning SAC
8 Kinodynamic Motion Planning for Mobile Robot Navigation across Inconsistent World Models 针对不一致环境模型的移动机器人运动规划方法GEGRH motion planning world model
9 Evolutionary Continuous Adaptive RL-Powered Co-Design for Humanoid Chin-Up Performance 提出EA-CoRL框架，解决人形机器人协同设计中控制策略对硬件的持续适应问题，提升引体向上性能。 humanoid humanoid robot reinforcement learning
10 Act to See, See to Act: Diffusion-Driven Perception-Action Interplay for Adaptive Policies 提出Action-Guided Diffusion Policy，通过扩散模型驱动感知-动作交互，提升策略自适应性。 manipulation policy learning imitation learning
11 TGPO: Temporal Grounded Policy Optimization for Signal Temporal Logic Tasks 提出TGPO，通过时序分解和分层优化解决基于STL的任务规划问题 quadruped locomotion manipulation ✅
12 RoboPilot: Generalizable Dynamic Robotic Manipulation with Dual-thinking Modes RoboPilot：双重思考模式实现通用动态机器人操作 manipulation chain-of-thought
13 Analytic Conditions for Differentiable Collision Detection in Trajectory Optimization 提出可微碰撞检测解析条件，加速轨迹优化中的非穿透约束 trajectory optimization penetration
14 The Trajectory Bundle Method: Unifying Sequential-Convex Programming and Sampling-Based Trajectory Optimization 提出轨迹束方法，统一序列凸规划与采样轨迹优化，解决无导数轨迹优化问题。 trajectory optimization motion planning
15 Best of Sim and Real: Decoupled Visuomotor Manipulation via Learning Control in Simulation and Perception in Real 提出解耦的视觉运动操作框架，提升Sim-to-Real迁移性能 manipulation sim-to-real
16 Anomaly detection for generic failure monitoring in robotic assembly, screwing and manipulation 提出基于自编码器的通用异常检测方法，用于机器人装配、拧螺丝和操作中的故障监控。 manipulation diffusion policy
17 On the Conic Complementarity of Planar Contacts 提出平面Signorini条件，统一离散与连续接触模型，扩展压力中心概念。 locomotion manipulation penetration
18 A Robotic Stirring Method with Trajectory Optimization and Adaptive Speed Control for Accurate Pest Counting in Water Traps 提出基于轨迹优化和自适应速度控制的机器人搅拌方法，用于水体陷阱中害虫的精确计数。 trajectory optimization
19 Hierarchical Diffusion Motion Planning with Task-Conditioned Uncertainty-Aware Priors 提出一种分层扩散运动规划器，利用任务条件的不确定性感知先验，提升规划效果。 motion planning
20 A Novel Robust Control Method Combining DNN-Based NMPC Approximation and PI Control: Application to Exoskeleton Squat Movements 提出基于DNN的NMPC近似与PI控制混合方法，提升外骨骼机器人下蹲运动的鲁棒性。 model predictive control
21 Learning from Hallucinating Critical Points for Navigation in Dynamic Environments 提出LfH-CP框架，通过幻觉关键点学习动态环境导航 motion planning
22 State Estimation for Compliant and Morphologically Adaptive Robots 针对柔顺和形态自适应机器人，提出一种基于神经网络的状态估计方法。 locomotion

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
23 dVLA: Diffusion Vision-Language-Action Model with Multimodal Chain-of-Thought 提出dVLA：基于扩散模型和多模态CoT的视觉-语言-动作机器人控制 vision-language-action VLA multimodal
24 LLM-MCoX: Large Language Model-based Multi-robot Coordinated Exploration and Search 提出LLM-MCoX以解决多机器人协同探索与搜索问题 large language model multimodal
25 TacRefineNet: Tactile-Only Grasp Refinement Between Arbitrary In-Hand Object Poses TacRefineNet：提出一种仅用触觉的多指灵巧手末端姿态精确调整方法 vision-language-action VLA

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
26 DiSA-IQL: Offline Reinforcement Learning for Robust Soft Robot Control under Distribution Shifts DiSA-IQL：面向分布偏移下柔性机器人鲁棒控制的离线强化学习 reinforcement learning deep reinforcement learning DRL
27 VLA Model Post-Training via Action-Chunked PPO and Self Behavior Cloning 提出基于动作块PPO和自行为克隆的VLA模型后训练方法 reinforcement learning PPO behavior cloning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
28 MUVLA: Learning to Explore Object Navigation via Map Understanding MUVLA：通过地图理解学习物体导航 semantic map vision-language-action
29 Terrain-Awared LiDAR-Inertial Odometry for Legged-Wheel Robots Based on Radial Basis Function Approximation 提出一种地形感知的激光雷达惯性里程计，用于腿轮机器人解决非结构化地形下的位姿漂移问题。 LIO

⬅️ 返回 cs.RO 首页 · 🏠 返回主页